254 Exp Physiol 93.

2 pp 254–261

Experimental Physiology

Magnesium sulphate treatment decreases blood–brain

barrier permeability during acute hypertension in
pregnant rats
Anna G. Euser, Lisa Bullinger and Marilyn J. Cipolla
Departments of Neurology, Obstetrics and Gynecology, and Pharmacology, University of Vermont College of Medicine, Burlington, VT 05405, USA

Eclampsia is associated with increased blood–brain barrier (BBB) permeability and formation of
cerebral oedema. Magnesium sulphate is used to treat eclampsia despite an unclear mechanism of
action. This study was to determine the effect of magnesium sulphate on in vivo BBB permeability
and formation of cerebral oedema during acute hypertension and on brain aquaporin-4 (AQP4)
protein expression. An in vivo model of hypertensive encephalopathy was used in late-pregnant
(LP) rats following magnesium sulphate treatment, 270 mg kg−1 I.P. injection every 4 h for 24 h.
Permeability of the BBB was determined by in situ brain perfusion of Evan’s Blue (EB) and sodium
fluorescein (NaFl), and dye clearance determined by fluorescence spectrophotometry. Cerebral
oedema was determined following acute hypertension by measuring brain water content. The
effect of magnesium treatment on AQP4 expression was determined by Western blot analysis.
Acute hypertension with autoregulatory breakthrough increased BBB permeability to EB in both
brain regions studied (P < 0.05). Magnesium attenuated BBB permeability to EB during acute
hypertension by 41% in the posterior cerebrum (P < 0.05) but had no effect in the anterior
cerebrum (P > 0.05). Treatment with magnesium did not change NaFl permeability, cerebral
oedema formation or AQP4 expression. In summary, BBB permeability to Evan’s Blue was
increased by acute hypertension in LP rats, and this was attenuated by treatment with magnesium
sulphate. The greatest effect on BBB permeability to EB was in the posterior cerebrum, an area
particularly susceptible to oedema formation during eclampsia.
(Received 15 August 2007; accepted after revision 2 October 2007; first published online 12 October 2007)
Corresponding author M. J. Cipolla: University of Vermont, Department of Neurology, 89 Beaumont Avenue, Given
C454, Burlington, VT 05405, USA. Email: marilyn.cipolla@uvm.edu

Eclampsia is a serious hypertensive disorder of pregnancy
associated with increased blood–brain barrier (BBB)
permeability and subsequent vasogenic oedema formation
(Schwartz et al. 1992, 2000; Engelter et al. 2000; Zeeman
et al. 2004). This condition is thought to be a form of
hypertensive encephalopathy (HTE; Schwartz et al. 1992,
2000; Easton, 1998), and both eclampsia and HTE are
causes of posterior reversible encephalopathy syndrome
(PRES; Hinchey et al. 1996; Lamy et al. 2004). Using a
model of HTE, we previously showed that acutely elevated
arterial pressure with breakthrough of cerebral blood
flow (CBF) autoregulation caused a significant increase
in cerebral oedema formation in late-pregnant (LP) rats,
which was not seen in non-pregnant (NP) control rats
despite similar pressures of autoregulatory breakthrough
(Euser & Cipolla, 2007). This suggests that pregnancy alone
promotes oedema formation under conditions of acute
hypertension.
Magnesium sulphate is widely used in North America
both to prevent and to treat eclamptic convulsions
(Witlin & Sibai, 1998). This treatment has been proven
to be more effective than anticonvulsant drugs and
placebo (The Eclampsia Trial Collaborative Group, 1995;
Altman et al. 2002), though the mechanism of action
remains unclear. Some studies suggest that magnesium
may prevent eclamptic seizures via vasodilatation in
the cerebral circulation (Belfort & Moise, 1992; Belfort
et al. 1993; Naidu et al. 1996). However, magnesium
treatment has also been reported to have little to no
effect on cerebral haemodynamics and CBF (Belfort
et al. 1999; Sherman et al. 2003; Hatab et al. 2005). For
example, we previously showed that while magnesium

DOI: 10.1113/expphysiol.2007.039966 
C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
Exp Physiol 93.2 pp 254–261 Magnesium sulphate and blood–brain barrier permeability
sulphate has a modest vasodilatory effect on cerebral
resistance arteries, the sensitivity of this response is
decreased by pregnancy and the postpartum state (Euser &
Cipolla, 2005). Furthermore, a randomized controlled trial
found that when compared with nimodipine, a calcium
channel blocker with specific cerebral vasodilator action,
magnesium sulphate was more effective in preventing
eclamptic seizures (Belfort et al. 2003). Together, these
results provide evidence that the primary action of
magnesium sulphate in eclampsia is likely not to be the
relief of cerebral vasospasm.
Treatment with magnesium sulphate has been reported
to decrease BBB permeability and cerebral oedema
formation in a variety of brain injury conditions, including
traumatic brain injury (Okiyama et al. 1995; Feldman
et al. 1996; Esen et al. 2003; Ghabriel et al. 2006), septic
encephalopathy (Esen et al. 2005), hypoglycaemia (Kaya
et al. 2001) and hyperosmolar mannitol injection (Kaya
et al. 2004). We therefore hypothesized that the action
of magnesium sulphate in the prophylaxis of eclampsia
may be related to protection of the BBB during acute
hypertension. The goal of this study was to determine
the effect of treatment with magnesium sulphate on in
vivo BBB permeability and formation of cerebral oedema
following CBF autoregulatory breakthrough in LP rats. In
addition, it has been proposed that magnesium sulphate
may limit cerebral oedema formation by decreasing the
expression of aquaporin-4 (AQP4; Ghabriel et al. 2006),
a water channel protein highly expressed in the brain,
although this interaction has not been directly shown.
Therefore, another goal of this study was to determine
the effect of treatment with magnesium sulphate on AQP4
protein expression in LP rats without acute hypertension.
Methods
Animals
All experiments used a rat model of pregnancy in
which primiparous Sprague–Dawley rats (Charles River,
St Constant, PQ, Canada) were studied on day 19–21
of a 22 day gestation. For each outcome measure, one
group of animals was treated with an i.p. injection of
magnesium sulphate (270 mg kg−1 every 4 h for 24 h)
prior to acute hypertension. This dosage of magnesium
sulphate has been reported to produce serum magnesium
levels in rats within the range (4.2–8.4 mg dl−1 , 0.35–
0.70 mmol l−1 ) recommended for eclamptic seizure
prophylaxis in pregnant women over 12–24 h (Pritchard,
1955; Hallak et al. 1994; Leveno & Cunningham, 1999). For
experiments investigating BBB permeability or cerebral
oedema formation in vivo, three groups of LP rats
were studied: animals without acute hypertension or
magnesium sulphate treatment (sham, n = 10); animals
that underwent acute hypertension alone (HTN, n = 7);

C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
255
and animals treated with magnesium sulphate prior to
acute hypertension (HTN + Mg2+ , n = 7). Sham rats did
not undergo acute hypertension, though experimental
length was comparable. For analysis of AQP4 protein
expression, separate groups of rats (n = 3 per group)
were used: LP, LP + Mg2+ and non-pregnant (NP). These
animals did not undergo acute hypertension, since we were
interested in the effect of magnesium treatment on AQP4
protein expression only. All animals were housed in the
University of Vermont Animal Care Facility, an American
Association for the Accrediatation of Laboratory Animal
Care accredited facility. All experimental procedures were
approved by the University of Vermont Institutional
Animal Care and Use Committee.
In vivo model of HTE
An in vivo model of HTE was used to determine
both BBB permeability and cerebral oedema formation
during acute hypertension, as previously described (Euser
& Cipolla, 2007). This model of HTE is considered
a model of eclampsia (Cipolla, 2007) and has been
used previously to study BBB permeability and oedema
formation during pregnancy (Euser & Cipolla, 2007).
Briefly, anaesthesia was initiated with isoflurane (≤ 3%
in oxygen, inhaled; Abbott, North Chicago, IL, USA)
and maintained with i.v. pentobarbital (≤ 60 mg kg−1 h−1 ;
Ovation Pharmaceuticals, Deerfield, IL, USA), which was
decreased during the surgical preparation, as tolerated,
to minimize the effects of anaesthesia on experimental
parameters. Adequate anaesthesia was assessed by toe
pinch and changes in arterial blood pressure. During
the course of each experiment, CBF was measured
transcranially using laser Doppler flowmetry with a
1.0 mm probe (Perimed, North Royalton, OH, USA)
affixed over a thinned area of skull (posterior to the coronal
suture and lateral to the sagittal suture) over the middle
cerebral artery perfusion domain, as described elsewhere
(Smeda et al. 1999). All laser Doppler measurements
of CBF were normalized to the flow at baseline (after
anaesthesia had been minimized and prior to acute
hypertension) to determine a relative CBF (rCBF). The
following equation was used:
rCBF = (CBFmax /CBFbaseline )
where CBFmax is the flow in laser Doppler units at maximal
pressure and CBFbaseline is the flow in laser Doppler units
prior to acute hypertension and after anaesthesia had been
minimized. Therefore, a rCBF of 2.0 signifies a twofold
increase in CBF from baseline.
Catheters were placed in the femoral artery and
veins for arterial blood pressure measurement and drug
delivery, respectively. Blood pressures were measured
via a pressure transducer attached in-line with the
256 A. G. Euser and others
catheter. Acute hypertension was induced by i.v. infusion
of phenylephrine (0.01 g per 10 ml lactated Ringer’s
solution; all reagents from Sigma, St Louis, MO, USA
unless otherwise specified). Following 10 min of blood
pressure 180 mmHg, sufficient to cause autoregulatory
breakthrough (Euser & Cipolla, 2007), the experiment
was ended with the humane death of the animal by
decapitation, and the brain was quickly removed. This
model of HTE was used with slight alterations to determine
both BBB permeability and cerebral oedema formation
during acute hypertension, described in the subsections
below.
Blood–brain barrier permeability studies
To determine the extent of BBB permeability, a lactated
Ringer solution (100 ml of solution, USP contains: NaCl
600 mg, NaC3 H5 O3 anhydrous 310 mg, KCl 30 mg and
CaCl2 dihydrate 20 mg. pH of the solution is 6.6 (6.0–
7.5).) Containing two different-sized dye tracers, 0.1%
sodium fluorescein (NaFl, 376 Da) and 2% Evan’s Blue
(EB, 69 kDa), was infused into the left ventricle of the
heart via catheter in the carotid artery. This solution was
allowed to circulate for 10 min prior to induction of acute
hypertension by i.v. infusion of phenylephrine (0.01 g
per 10 ml lactated Ringer solution). Following 10 min
of blood pressure 180 mmHg, again sufficient to cause
autoregulatory breakthrough (Euser & Cipolla, 2007),
the animal was perfused with lactated Ringer solution
through the central catheter in order to flush the dye
from the cerebral circulation. Following decapitation, the
brain was quickly removed, sectioned and weighed. The
cerebral cortices were separated from the cerebellum and
brainstem, and then further divided into anterior and
posterior cerebrum sections by a coronal cut at the level
of the optic chiasm. We chose to compare anterior and
posterior cerebrums because of the propensity of oedema
to form in the posterior brain regions during eclampsia
and PRES (Schwartz et al. 1992, 2000; Hinchey et al. 1996;
Engelter et al. 2000; Lamy et al. 2004; Zeeman et al. 2004;
Cipolla, 2007). Any animal in which gross observation
revealed inadequate flushing of the cerebrovasculature
was excluded from analysis (this occurred in 2 animals).
Samples were processed as previously described (Euser &
Cipolla, 2007), and the resulting supernatant was analysed
by fluorescence spectrophotometry at 460–515 nm for
NaFl and 620–680 nm for EB. Data are expressed as average
fluorescence counts per second (CPS) per gram brain
tissue.
Determination of cerebral oedema
The percentage water content of the brain is a measure of
cerebral oedema (Schwab et al. 1997). After elevated blood
pressure had been maintained for 10 min, animals were
Exp Physiol 93.2 pp 254–261
quickly decapitated and the brain was then removed for
wet and dry weight measurements. The brain was weighed
wet immediately after removal and then dried at 100◦ C
for 24 h, at which point the brain was weighed again dry.
Brain water content (as a percentage) was determined by
the following formula:
Brain water content = [(weightwet − weightdry )
÷ weightwet ] × 100%
where weightwet is the weight of the brain immediately after
removal from the animal and weightdry is the weight of the
brain after drying.
Western blot analysis of AQP4 expression
Following induction of anaesthesia with isoflurane
(Abbott, North Chicago, IL, USA) and decapitation,
brains were carefully removed and divided into
anterior and posterior cerebrum, as described above.
Brain sections were snap frozen in liquid nitrogen
and stored at −80◦ C. For protein extraction, each
section was homogenized in a glass Dounce tissue
grinder with 3 ml lysis buffer comprising: 50 mm
Trizma® hydrochloride, 150 mm NaCl, 10 mm EDTA,
0.25% deoxycholate, 1% nonylphenyl polyethylene glycol
detergent (Calbiochem, San Diego, CA, USA), 10%
glycerol, 1% sodium dodecyl sulphate, 1 mm dithiothreitol
(Bio-Rad, Richmond, CA, USA) and 1% protease inhibitor
cocktail. The homogenate was transferred and centrifuged
at 3900 g for 10 min at 4◦ C. The supernatant was
centrifuged again under the same conditions. The total
amount of protein was measured using the Coomassie
Plus-BradfordTM . Assay Kit (Pierce, Rockford, IL, USA).
Protein samples were incubated in Laemmli sample
buffer (Bio-Rad) with 2β-mercaptoethanol at 95◦ C
for 10 min. Protein (10 µg) was separated by sodium
dodecyl sulphate-polyacrylamide gel electrophoresis and
transferred to a polyvinylidene difluoride membrane
(Bio-Rad). Membranes were blocked for 20 min at room
temperature in 3% non-fat milk in phosphate-buffered
saline containing 0.005% Tween-20 (PBST; Calbiochem),
cut vertically, and subsequently incubated overnight at
4◦ C with two primary antibodies: an affinity purified
rabbit polyclonal antibody raised against residues 249–
323 of rat aquaporin-4, 1:1000 (Chemicon, Temecula,
CA, USA), and a mouse monoclonal antibody to
glyceraldehyde-3-phosphate dehydrogenase (GAPDH),
1:30 000 (Biodesign, Saco, ME, USA). After repeated
washing in PBST, membranes were incubated in secondary
antibodies conjugated to horseradish peroxidase for 1 h
at room temperature. A sheep antirabbit IgG (Abcam,
Cambridge, MA, USA), 1:2000, was used for AQP4
and a goat antimouse IgG, 1:3000 (Pierce), was used
for GAPDH. Additional washing steps in PBST were

C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
Exp Physiol 93.2 pp 254–261 Magnesium sulphate and blood–brain barrier permeability
Table 1. Physiological characteristics of animals studied
Weight (g)
Sham (n = 10) 344 ± 4
HTN (n = 7) 336 ± 11
HTN + Mg2+ (n = 7) 342 ± 15
Abbreviations: ABP, arterial blood pressure; rCBF, relative cerebral blood flow; and HTN, acute hypertension.
sham group within same experimental conditions; †P < 0.05 versus sham and HTN.
followed by chemiluminescence using SuperSignal West
Pico Substrate and CL-XPosure Film (Pierce). Films were
scanned into Adobe Photoshop CS, and densitometry
of AQP4 and GAPDH bands was determined with
ImageJ software (version 1.37v, available for download
at http://rsb.info.nih.gov/ij/). All experiments were done
in duplicate. The ratio of AQP4/GAPDH intensity was
calculated for each animal in each group. Values for NP
animals were assigned a value of 1.0 and LP and LP + Mg2+
values were normalized to the NP group before analysing
each group.
Statistical analyses
All data are expressed as means ± s.e.m. Differences in
animal characteristics, tissue fluorescence and brain water
content between treatment groups were determined by
ANOVA with a post hoc Student–Newman–Keuls test for
multiple comparisons. Differences in tissue fluorescence
between the anterior and posterior cerebrum within the
same treatment group were determined by Student’s paired
5000
Sham
4000
HTN
HTN + Mg
Fluorescence (CPS/g)
3000
* *†
2000
1000
0
Anterior
Cerebrum
Brain Region
Figure 1. Bar graph showing average fluorescence (in CPS g−1 )
of Evan’s Blue in the anterior and posterior cerebrum of all
groups of late-pregnant animals as a measure of BBB
permeability
Blood–brain barrier permeability was significantly increased with acute
hypertension and autoregulatory breakthrough (HTN) compared with
∗
sham control animals ( P < 0.05). The increase in BBB permeability
was greater in the posterior versus anterior cerebrum (‡P < 0.05).
Magnesium sulphate treatment (HTN + Mg2+ ) significantly decreased
BBB permeability in the posterior cerebrum in response to acute
hypertension (†P < 0.05 versus HTN).

C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
257
Litter size Baseline ABP (mmHg) Maximal ABP (mmHg) Maximal rCBF
13 ± 1 118 ± 5 126 ± 5 1.07 ± 0.027
12 ± 1 116 ± 4 191 ± 3∗∗ 2.41 ± 0.29∗∗
12 ± 1 102 ± 3† 186 ± 4∗∗ 2.21 ± 0.30∗∗
∗∗ P < 0.01 versus
t test. Likewise, differences between baseline and maximal
blood pressures within the same treatment group were
determined by Student’s paired t test. Differences in AQP4
expression between LP and LP + Mg2+ groups were tested
by non-parametric Mann–Whitney U test. Differences
were considered significant if P < 0.05.
Results
The physiological characteristics of the LP animals studied
with the in vivo model of HTE are summarized in
Table 1. Two animals were excluded from study, one from
each group of HTN and HTN + Mg2+ animals, owing
to insufficient flushing of the dye from the vasculature.
There were no significant differences between treatment
groups in either animal weight or litter size. Baseline
mean arterial blood pressure, measured directly by femoral
arterial catheter, was not significantly different between
sham and HTN groups, but was significantly lower in
animals treated with magnesium (P < 0.05). Infusion
of phenylephrine to induce autoregulatory breakthrough
caused a significant increase in arterial blood pressure in
the HTN and HTN + Mg2+ groups compared with the
sham group (P < 0.01). The increase in blood pressure
was accompanied by a significant increase in CBF, as
represented by the rCBF ≥ 2.0 for both hypertensive
*‡ groups, indicating autoregulaotry breakthrough (P < 0.01
versus sham).
Acute hypertension with autoregulatory breakthrough
significantly increased BBB permeability to EB in both
the anterior and posterior cerebrum (Fig. 1). Interestingly,
the increase in BBB permeability was significantly greater
in the posterior brain region versus anterior (659 versus
365%, respectively, P < 0.05), a region most susceptible
to oedema formation. Magnesium sulphate treatment
Posterior
Cerebrum
decreased BBB permeability significantly only in the
posterior brain region (Fig. 1). There was a 41% decrease
in permeability in the posterior cerebrum (P < 0.05). The
BBB permeability to NaFl was unaffected by hypertension
or magnesium sulphate treatment (Fig. 2). The lack of a
significant difference in NaFl permeability probably results
from the higher variability in those groups compared with
EB.
Brain water content, a measure of cerebral oedema
formation (Schwab et al. 1997), was not different between
any of the groups. The percentage water content was:
78.25 ± 0.18% for sham group, 78.13 ± 0.02% for LP
258 A. G. Euser and others Exp Physiol 93.2 pp 254–261

50 Sham HTN group and 78.02 ± 0.07% for LP HTN + Mg2+

HTN group (n.s.). In contrast to our previous findings (Euser
HTN + Mg & Cipolla, 2007), there was no significant difference in
40
Fluorescence (CPS/g) x 1,000

cerebral oedema formation between any of the groups.

Aquaporin-4 protein expression was determined in
30 the anterior and posterior cerebrum of normotensive LP
and LP + Mg2+ groups and compared with NP control
20 animals. Expression of AQP4 was increased in LP and
LP + Mg2+ compared with NP animals in both the
anterior and posterior cerebrum by 3- to 4-fold (Fig. 3),
10
a finding similar to previous results (Quick & Cipolla,
2005). Contrary to previous reports (Ghabriel et al. 2006),
0 magnesium sulphate treatment did not have any effect on
Anterior Posterior
Cerebrum Cerebrum AQP4 expression.
Brain Region

Figure 2. Bar graph showing average fluorescence (in CPS g−1 )
of sodium fluorescein in the anterior and posterior cerebrum in
all groups of late-pregnant rats as a measure of BBB
permeability
There was no difference in BBB between animals that underwent
acute hypertension (HTN) compared with sham control animals.
Magnesium sulphate treatment (HTN + Mg2+ ) also had no effect on
BBB permeability to NaFl compared with HTN.
Discussion
The major finding of this study was that magnesium
sulphate treatment significantly decreased BBB
permeability to EB in LP rats during acute hypertension,
an effect that was significant in the posterior brain
region. In fact, BBB permeability to EB varied regionally

Figure 3. Aquaporin-4 (AQP4) expression in the female rat brain with and without magnesium
treatment
Representative Western blots of AQP4 expression in late-pregnant (LP) rats in the anterior cerebrum (A) and
posterior cerebrum (B). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) blots from the same gel were used
as an internal control. C shows the intensity ratios for untreated (LP) and treated animals (LP + Mg2+ ) compared
with non-pregnant control animals (NP). Aquaporin-4 expression was increased in both LP and LP + Mg2+ animals
compared with NP control animals, but was unchanged by magnesium sulphate treatment.


C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
Exp Physiol 93.2 pp 254–261 Magnesium sulphate and blood–brain barrier permeability
in response to autoregulatory breakthrough, such that
the posterior cerebrum showed a greater increase in
permeability than the anterior cerebrum. Since the
posterior brain region is most susceptible to oedema
formation during PRES and eclampsia (Schwartz et al.
1992, 2000; Hinchey et al. 1996; Engelter et al. 2000;
Lamy et al. 2004; Zeeman et al. 2004; Cipolla, 2007), these
results suggest that one mechanism by which oedema
forms preferentially in this region is the enhanced BBB
permeability to large proteins such as EB. Together, these
results suggest that the beneficial effect of magnesium
sulphate treatment during eclampsia may be related to
protection of the BBB that limits solute flux into the
brain tissue when hydrostatic pressure is elevated to
pathological levels.
The mechanism by which magnesium sulphate acts
to decrease BBB permeability is not clear from this
study, but it may be related to a direct effect on the
cerebral endothelium. Magnesium sulphate is a calcium
antagonist (Fawcett et al. 1999) and may decrease
paracellular transport through tight junctions by opposing
calcium-induced contractions of the actin cytoskeleton
in endothelial cells. Alternatively, acute hypertension has
been shown to increase pinocytosis that may enhance
transcellular transport in the cerebral endothelium (Nag
et al. 1977, 1979; Westergaard et al. 1977). Magnesium
may counter this mode of transport and decrease BBB
permeability during acute hypertension. In fact, it has
been suggested that transport of large molecules across the
BBB implicates transcellular versus paracellular transport
(Mayhan & Heistad, 1985). Since magnesium treatment
had a greater effect on a large solute (EB) than on a small
solute (NaFl), this suggests that transcellular transport
was the primary mode of BBB permeability under these
conditions.
Another protective effect of magnesium sulphate on
limiting BBB permeability during acute hypertension may
be related to its ability to scavenge free radicals. Free
radical production during acute hypertension induced by
pressor agents has been demonstrated (Zhang & Ellis,
1991). In fact, superoxide dismutase, an oxygen free radical
scavenger, was shown to reduce BBB permeability and
cerebral oedema during acute hypertension in rats (Zhang
& Ellis, 1990). It is therefore possible that magnesium
sulphate acts similarly to scavenge free radicals and protect
the BBB, since treatment with magnesium sulphate has
been shown to decrease lipid peroxidation in pre-eclamptic
women, suggesting that it has free radical scavenging
properties (Ariza et al. 2005).
In contrast to a previous study (Euser & Cipolla, 2007),
we did not find that cerebral oedema formation was
increased in response to acute hypertension in LP rats, or
that magnesium sulphate treatment affected brain water
content. This negative and contrary finding may be for
several reasons. First, the change in rCBF was considerably

C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
259
less in the present study compared with what we have
published previously. The purpose of the previous study
(Euser & Cipolla, 2007) was to determine the pressure
at which autoregulatory breakthrough occurred between
NP and LP rats, and those animals routinely had rCBF
> 3.0. This difference would make hydrostatic pressure
less in the present group of animals compared with
the previous study. Hydrostatic pressure is a primary
determinant of brain oedema (Shima, 2003). Therefore,
a lower rCBF would suggest a lower hydrostatic pressure
and could account for the undetectable changes in brain
water content in the present study. It is also possible that
the duration of acute hypertension, being only 10 min,
was not sufficient to cause a detectable increase in oedema
formation with this degree of rCBF change. While this
methodology has been commonly used to study BBB
permeability in response to acute hypertension, it probably
takes some time for oedema to form. Unfortunately, this
non-survival model of HTE does not allow pressure to be
increased for longer durations (probably due to peripheral
vascular dilatation that decreases total peripheral vascular
resistance and drops arterial blood pressure). Other
methods in which the animals survive for longer periods,
such as 2–3 days, would be better for assessment of
oedema in response to acute hypertension and magnesium
sulphate treatment.
Previous studies have suggested that magnesium
sulphate treatment may limit cerebral oedema formation
by decreasing the expression of AQP4 (Ghabriel et al.
2006). Aquaporin-4 is a transmembrane water channel
highly expressed in the brain and shown to be upregulated
during pregnancy (Quick & Cipolla, 2005). Owing
to its water fluxing capabilities (Papadopoulos et al.
2004), a decrease induced by magnesium may limit
oedema formation. For this reason, we investigated
whether magnesium sulphate treatment affected AQP4
protein expression. Animals were treated with magnesium
sulphate for 24 h, a duration consistent with clinical
practice, but perhaps not long enough to have an effect
on AQP4 protein levels. Figure 3 shows that AQP4 was
increased in both pregnant groups compared with NP
animals, but magnesium sulphate treatment had no effect
on its expression. These measurements were done on
normotensive animals because we wanted to determine
whether there was any effect of magnesium treatment prior
to a hypertensive insult that may limit oedema formation.
How AQP4 protein expression changes in response to
hypertension is of clear interest; however, animals would
probably need to survive for longer time periods.
This study used a model of HTE during pregnancy
to investigate how acute hypertension affects BBB
permeability. This model has been extensively used
to investigate the effect of cerebral haemodynamics
on BBB permeability and vasogenic oedema formation
(MacKenzie et al. 1976; Baumbach & Heistad, 1985;
260 A. G. Euser and others
Hatashita et al. 1985, 1986; Mayhan & Heistad, 1985;
Mayhan et al. 1986; Euser & Cipolla, 2007) and is
considered a model of eclampsia (Cipolla, 2007). It is worth
noting, however, that we studied the effect of magnesium
treatment on BBB permeability during normal pregnancy,
not under conditions of oxidative stress or endothelial cell
damage which may have a role in eclampsia. However,
this model is valuable and has a number of the features
relevant to the pathogenesis of eclampsia, including loss
of autoregulation, hyperperfusion and BBB disruption. It
is also useful for evaluating regional differences in BBB
permeability, which are an important feature considering
the regional heterogeneity in oedema formation during
eclampsia.
To our knowledge, this is the first study to investigate
the effect of magnesium sulphate treatment on BBB
permeability during acute hypertension in pregnancy.
The results demonstrate that treatment with magnesium
sulphate decreased BBB permeability following acute
hypertension, particularly in the posterior cerebrum.
However, magnesium treatment did not affect cerebral
oedema formation during the short time course studied. A
more complete understanding of the effect of magnesium
sulphate on the BBB may provide information regarding
the beneficial effect of magnesium in the treatment and
prophylaxis of eclampsia.
References
Altman D, Carroli G, Duley L, Farrell B, Moodley J, Neilson J,
Smith D & Magpie Trial Collaboration Group (2002). Do
women with pre-eclampsia, and their babies, benefit from
magnesium sulphate? The Magpie Trial: a randomised
placebo-controlled trial. Lancet 359, 1877–1890.
Ariza AC, Bobadilla N, Fernandez C, Munoz-Fuentes RM,
Larrea F & Halhali A (2005). Effects of magnesium sulfate on
lipid peroxidation and blood pressure regulators in
preeclampsia. Clin Biochem 38, 128–133.
Baumbach GL & Heistad DD (1985). Heterogeneity of brain
blood flow and permeability during acute hypertension.
Am J Physiol Heart Circ Physiol 249, H629–H637.
Belfort MA, Anthony J, Saade GR, Allen JC Jr & Nimodipine
Study Group (2003). A comparison of magnesium sulfate
and nimodipine for the prevention of eclampsia.
N Engl J Med 348, 304–311.
Belfort MA & Moise KJ Jr (1992). Effect of magnesium sulfate
on maternal brain blood flow in preeclampsia: a randomized,
placebo-controlled study. Am J Obstet Gynecol 167, 661–666.
Belfort MA, Saade GR & Moise KJ Jr (1993). The effect of
magnesium sulfate on maternal and fetal blood flow in
pregnancy-induced hypertension. Acta Obstet Gynecol Scand
72, 526–530.
Belfort MA, Saade GR, Yared M, Grunewald C, Herd JA, Varner
MA & Nisell H (1999). Change in estimated cerebral
perfusion pressure after treatment with nimodipine or
magnesium sulfate in patients with preeclampsia.
Am J Obstet Gynecol 181, 402–407.
Exp Physiol 93.2 pp 254–261
Cipolla MJ (2007). Cerebrovascular changes in pregnancy and
eclampsia. Hypertension 50, 14–24.
Easton JD (1998). Severe preeclampsia/eclampsia: hypertensive
encephalopathy of pregnancy? Cerebrovasc Dis 8, 53–58.
Engelter ST, Provenzale JM & Petrella JR (2000). Assessment of
vasogenic edema in eclampsia using diffusion imaging.
Neuroradiology 42, 818–820.
Esen F, Erdem T, Aktan D, Kalayci R, Cakar N, Kaya M & Telci
L (2003). Effects of magnesium administration on brain
edema and blood–brain barrier breakdown after
experimental traumatic brain injury in rats. J Neurosurg
Anesthesiol 15, 119–125.
Esen F, Erdem T, Aktan D, Orhan M, Kaya M, Eraksoy H,
Cakar N & Telci L (2005). Effect of magnesium sulfate
administration on blood–brain barrier in a rat model of
intraperitoneal sepsis: a randomized controlled experimental
study. Crit Care 9, R18–R23.
Euser AG & Cipolla MJ (2005). Resistance artery vasodilation
to magnesium sulfate during pregnancy and the postpartum
state. Am J Physiol Heart Circ Physiol 288, H1521–H1525.
Euser AG & Cipolla MJ (2007). Cerebral blood flow
autoregulation and edema formation during pregnancy in
anesthetized rats. Hypertension 49, 334–340.
Fawcett WJ, Haxby EJ & Male DA (1999). Magnesium:
physiology and pharmacology. Br J Anaesth 83, 302–320.
Feldman Z, Gurevitch B, Artru AA, Oppenheim A, Shohami E,
Reichenthal E & Shapira Y (1996). Effect of magnesium
given 1 hour after head trauma on brain edema and
neurological outcome. J Neurosurg 85, 131–137.
Ghabriel MN, Thomas A & Vink R (2006). Magnesium restores
altered aquaporin-4 immunoreactivity following traumatic
brain injury to a pre-injury state. Acta Neurochir Suppl 96,
402–406.
Hallak M, Berman RF, Irtenkauf SM, Janusz C & Cotton DB
(1994). Magnesium sulfate treatment decreases N -methyl-
D-aspartate receptor binding in the rat brain: an
autoradiographic study. J Soc Gynecol Invest 1, 25–30.
Hatab MR, Zeeman GG & Twickler DM (2005). The effect
of magnesium sulfate on large cerebral artery blood flow
in severe preeclampsia. J Matern Fetal Neonat Med 17,
187–192.
Hatashita S, Hoff JT & Ishii S (1986). Focal brain edema
associated with acute arterial hypertension. J Neurosurg 64,
643–649.
Hatashita S, Koike J, Sonokawa T & Ishii S (1985). Cerebral
edema associated with craniectomy and arterial
hypertension. Stroke 16, 661–668.
Hinchey J, Chaves C, Appignani B, Breen J, Pao L, Wang A,
Pessin MS, Lamy C, Mas JL & Caplan LR (1996). A reversible
posterior leukoencephalopathy syndrome. N Engl J Med 334,
494–500.
Kaya M, Gulturk S, Elmas I, Arican N, Kocyildiz ZC, Kucuk M,
Yorulmaz H & Sivas A (2004). The effects of magnesium
sulfate on blood–brain barrier disruption caused by
intracarotid injection of hyperosmolar mannitol in rats. Life
Sci 76, 201–212.
Kaya M, Kucuk M, Kalayci RB, Cimen V, Gurses C, Elmas I &
Arican N (2001). Magnesium sulfate attenuates increased
blood–brain barrier permeability during insulin-induced
hypoglycemia in rats. Can J Physiol Pharmacol 79, 793–798.

C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
Exp Physiol 93.2 pp 254–261 Magnesium sulphate and blood–brain barrier permeability
Lamy C, Oppenheim C, Meder JF & Mas JL (2004).
Neuroimaging in posterior reversible encephalopathy
syndrome. J Neuroimaging 14, 89–96.
Leveno KJ & Cunningham FG (1999). Management of
preeclampsia. In Chesley’s Hypertensive Disorders in
Pregnancy, ed. Lindheimer MD, Roberts JM & Cunningham
FG, pp. 543–580. Appleton & Lange, Stamford, CT, USA.
MacKenzie ET, Strandgaard S, Graham DI, Jones JV & Harper
AM (1976). Effects of acutely induced hypertension in cats
on pial arteriolar caliber, local cerebral blood flow, and the
blood–brain barrier. Circ Res 39, 33–41.
Mayhan WG, Faraci FM & Heistad DD (1986). Disruption of
the blood–brain barrier in cerebrum and brain stem during
acute hypertension. Am J Physiol Heart Circ Physiol 251,
H1171–H1175.
Mayhan WG & Heistad DD (1985). Permeability of blood–
brain barrier to various sized molecules. Am J Physiol Heart
Circ Physiol 248, H712–H718.
Nag S, Robertson DM & Dinsdale HB (1977). Cerebral cortical
changes in acute experimental hypertension: an
ultrastructural study. Lab Invest 36, 150–161.
Nag S, Robertson DM & Dinsdale HB (1979). Quantitative
estimate of pinocytosis in experimental acute hypertension.
Acta Neuropathol (Berl) 46, 107–116.
Naidu S, Payne AJ, Moodley J, Hoffmann M & Gouws E (1996).
Randomised study assessing the effect of phenytoin and
magnesium sulphate on maternal cerebral circulation in
eclampsia using transcranial Doppler ultrasound. Br J Obstet
Gynaecol 103, 111–116.
Okiyama K, Smith DH, Gennarelli TA, Simon RP, Leach M &
McIntosh TK (1995). The sodium channel blocker and
glutamate release inhibitor BW1003C87 and magnesium
attenuate regional cerebral edema following experimental
brain injury in the rat. J Neurochem 64, 802–809.
Papadopoulos MC, Manley GT, Krishna S & Verkman AS
(2004). Aquaporin-4 facilitates reabsorption of excess fluid
in vasogenic brain edema. FASEB J 18, 1291–1293.
Pritchard JA (1955). The use of the magnesium ion in the
management of eclamptogenic toxemias. Surg Gynecol Obstet
100, 131–140.
Quick AM & Cipolla MJ (2005). Pregnancy-induced
up-regulation of aquaporin-4 protein in brain and its role in
eclampsia. FASEB J 19, 170–175.
Schwab M, Bauer R & Zwiener U (1997). The distribution of
normal brain water content in Wistar rats and its increase
due to ischemia. Brain Res 749, 82–87.
Schwartz RB, Feske SK, Polak JF, DeGirolami U, Iaia A, Beckner
KM, Bravo SM, Klufas RA, Chai RY & Repke JT (2000).
Preeclampsia-eclampsia: clinical and neuroradiographic
correlates and insights into the pathogenesis of hypertensive
encephalopathy. Radiology 217, 371–376.

C 2007 The Authors. Journal compilation 
C 2007 The Physiological Society
261
Schwartz RB, Jones KM, Kalina P, Bajakian RL, Mantello
MT, Garada B & Holman BL (1992). Hypertensive
encephalopathy: findings on CT, MR imaging, and
SPECT imaging in 14 cases. AJR Am J Roentgenol 159,
379–383.
Sherman R, Armory P, Moody P, Hope T & Mahajan RP
(2003). Effects of magnesium sulphate on cerebral
haemodynamics in healthy volunteers: a transcranial
Doppler study. Br J Anaesth 91, 273–275.
Shima K (2003). Hydrostatic brain edema: basic mechanisms
and clinical aspect. Acta Neurochir Suppl 86, 17–20.
Smeda JS, VanVliet BN & King SR (1999). Stroke-prone
spontaneously hypertensive rats lose their ability to
auto-regulate cerebral blood flow prior to stroke. J Hypertens
17, 1697–1705.
The Eclampsia Trial Collaborative Group(1995). Which
anticonvulsant for women with eclampsia? Evidence from
the Collaborative Eclampsia Trial. Lancet 345, 1455–1463.
Westergaard E, van Deurs B & Bronsted HE (1977). Increased
vesicular transfer of horseradish peroxidase across cerebral
endothelium, evoked by acute hypertension. Acta
Neuropathol (Berl) 37, 141–152.
Witlin AG & Sibai BM (1998). Magnesium sulfate therapy in
preeclampsia and eclampsia. Obstet Gynecol 92, 883–889.
Zeeman GG, Fleckenstein JL, Twickler DM & Cunningham FG
(2004). Cerebral infarction in eclampsia. Am J Obstet Gynecol
190, 714–720.
Zhang XM & Ellis EF (1990). Superoxide dismutase reduces
permeability and edema induced by hypertension in rats.
Am J Physiol Heart Circ Physiol 259, H497–H503.
Zhang XM & Ellis EF (1991). Superoxide dismutase decreases
mortality, blood pressure, and cerebral blood flow responses
induced by acute hypertension in rats. Stroke 22, 489–494.
Acknowledgements
We gratefully acknowledge the support of the American
Heart Association Established Investigator Award (0540081N
to M.J.C.), the American Heart Association North-east Affiliate
Research Committee Predoctoral Fellowship (000019871 to
A.G.E.), the National Institute of Neurological Disorders and
Stroke (R01 NS-045940 to M.J.C.), the Totman Medical Research
Trust, and the University of Vermont College of Medicine
MD/PhD Program.